Low-voltage transformer area branch relation checking method and device based on inter-user voltage record spectral clustering

Abstract

The invention discloses a low-voltage transformer area branch relation checking method based on inter-user voltage record spectral clustering. The method comprises the following steps: extracting the moment when the voltage effective value of each user in the same transformer area is reduced and increased; counting a rising and falling matching number between the voltage effective value cross phases of each to-be-checked user; constructing a voltage incidence matrix according to the normalized matching coefficient; constructing user samples according to the voltage incidence matrix for spectral clustering; and comparing a clustering result with a system file to perform branch relationship checking. According to the method, the voltage change rule between the users is analyzed, grouping and topological relation inference are conducted on the users through the spectral clustering algorithm, the problems that in a traditional method, granularity is insufficient, noise interference is large, and real-time performance is low can be effectively solved, the recognition effect is stable, adaptability is high, meanwhile, the method is suitable for a complex low-voltage distribution network structure, and practicability is high. And branch relation identification requirements in different scenes can be met.

Description

Technical Field

[0001] The present invention belongs to the technical field of power systems, and particularly relates to a method for checking the branch relationship of low-voltage substations. Background Art

[0002] Checking the branch relationship of low-voltage substations is a key task in distribution network management, aiming to accurately identify and check the topological relationship among user meters, meter boxes, and branch boxes in low-voltage substations. The accurate identification of branch relationships is directly related to the accuracy of distribution network operation monitoring, the effectiveness of power consumption information acquisition systems, and the reliability of power supply services. With the continuous expansion of the scale of low-voltage distribution networks, their network structures are becoming increasingly complex, and checking branch relationships faces many challenges.

[0003] Currently, low-voltage distribution networks are usually arranged in a tree or ring network structure, and power distribution to low-voltage users is achieved through branch boxes and meter boxes. However, due to the diverse wiring methods and dynamically changing operating environments of low-voltage distribution networks, traditional branch relationship identification methods have certain limitations in practical applications. Existing methods mainly rely on the analysis of power parameters such as voltage and current, combined with geographic information systems (GIS) for spatial data processing. These methods have certain effects under the conditions of simple topology and stable operating states of low-voltage distribution networks, but when dealing with complex branch network structures, they face the following problems: 1. Insufficient data acquisition granularity: Most existing power consumption information acquisition systems use user meters as data sources, lacking accurate data support at the levels of branch boxes and meter boxes, and it is difficult to achieve accurate branch relationship checking through direct analysis. This problem mainly stems from limited equipment coverage, high installation costs, and constraints on data acquisition and transmission technologies. 2. Accuracy problems under complex structures: The branch structures of low-voltage distribution networks are diverse and dynamic. Especially when there are multiple user accesses or load fluctuations, traditional analysis methods are easily interfered by noise data, resulting in a reduced accuracy rate of branch relationship identification. In addition, the similarity of electrical parameters between users further increases the difficulty of branch relationship identification. 3. Lack of real-time performance: Traditional methods are mostly based on rules or static models and are difficult to adapt to the dynamically changing operating environments of low-voltage distribution networks. This leads to the update and maintenance of branch relationships often lagging behind the actual operating conditions in practical applications, thus affecting power supply reliability and fault location efficiency. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method for checking the branch relationship of low-voltage substations based on voltage record spectrum clustering among users. This method analyzes the voltage change laws among users and uses the spectrum clustering algorithm to group users and infer topological relationships, which can effectively overcome the problems of insufficient granularity, large noise interference, and low real-time performance in traditional methods, and is also applicable to complex low-voltage distribution network structures.

[0005] The technical solution adopted by the present invention is as follows: A method for verifying the branch relationship of a low-voltage substation based on the clustering of voltage recording spectra among users, including:

[0006] Extract the moments when the effective voltage values of each user in the same substation decrease and increase;

[0007] According to the extracted moments, count the number of up-down matches between the effective voltage values of each user to be verified across phases;

[0008] Construct a voltage correlation matrix based on the normalized matching coefficient and the number of matches;

[0009] Construct user samples according to the voltage correlation matrix for spectral clustering;

[0010] Compare the clustering results with the system archives to verify the branch relationship.

[0011] Extract the moments when the effective voltage values of each user in the same substation decrease and increase, specifically:

[0012] According to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value within 0.1 second is greater than or equal to 0.1 volt, forming a record of the voltage change moments of the user, including: the time point that meets the conditions and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour system.

[0013] According to the extracted moments, count the number of up-down matches between the effective voltage values of each user to be verified across phases, specifically: Select all the voltage change moment records of each user to be verified on the same day. Record the record of the increase in phase B as Up.k, and record the record of the decrease in phase A as Down.k, where k represents that the record data at this moment is the kth user's, and A, B, C are the default phase sequences; num(Up.i→Down.j) is the number of matches between the increase in phase B and the decrease in phase A of users i and j; num(Up.i→Down.i) is the number of matches between the increase in phase B and the decrease in phase A of user i itself; the num(*) function is used to count the number of elements in the record set.

[0014] Construct a voltage correlation matrix based on the normalized matching coefficient and the number of matches, specifically:

[0015] Construct an n×n matrix M according to the number n of users whose branch relationship needs to be verified. All initial elements of M are 0;

[0016] Assign different numbers to the users to be verified, numbered 1, 2, 3,..., n;

[0017] Assign values to the elements in the voltage correlation matrix M according to the normalized matching coefficient, specifically That is, the ratio of the number of up-down matches between user i and user j to the number of matches of user i itself.

[0018] Construct user samples based on the voltage incidence matrix for spectral clustering, specifically as follows:

[0019] Take each column of data in M as a sample to construct the degree matrix D:

[0020] Construct the Laplacian matrix L based on the degree matrix D: L = D - M;

[0021] Perform eigenvalue decomposition on the Laplacian matrix L, and take the eigenvectors corresponding to the first K smallest eigenvalues to form the matrix V, where K is the number of branches;

[0022] Perform K-means clustering on each row of the matrix V to divide the users into K clusters.

[0023] The present invention also provides a low-voltage substation branch relationship verification device based on spectral clustering of voltage records between users, including:

[0024] A voltage change time extraction module for extracting the times when the effective voltage values of each user in the same substation decrease and increase;

[0025] A matching number statistics module for statistically calculating the number of up-down matches between the effective voltage values of each user to be verified across phases according to the extracted times;

[0026] An incidence matrix construction module for constructing a voltage incidence matrix based on the normalized matching coefficient according to the matching number;

[0027] A clustering module for constructing user samples based on the voltage incidence matrix for spectral clustering;

[0028] A verification module for verifying the branch relationship by comparing the clustering results with the system archives.

[0029] The voltage change time extraction module is specifically used for: according to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation, extracting and recording the corresponding times when the absolute value of the decrease and increase of its effective value within 0.1 second is greater than or equal to 0.1 volt, forming a voltage change time record of the user, and the record includes: the time point that meets the condition and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format.

[0030] The matching number statistics module is specifically used for: selecting all voltage change moment records of each user to be verified on the same day, recording the records of the rising of phase B as Up.k, and recording the records of the falling of phase A as Down.k, where k represents that the record data at this moment is the k-th user's, and A, B, and C are the default phase sequences; num(Up.i→Down.j) is the matching number of the rising of phase B and the falling of phase A between user i and user j; num(Up.i→Down.i) is the matching number of the rising of phase B and the falling of phase A of user i itself; the num(*) function is used to count the number of elements in the record set.

[0031] The correlation matrix construction module is specifically used for:

[0032] Construct an n×n matrix M according to the number n of users whose branch relationships need to be verified. All initial elements of M are 0;

[0033] Assign different numbers to the users to be verified, and the numbers are 1, 2, 3,..., n;

[0034] Assign values to the elements in the voltage correlation matrix M according to the normalized matching coefficient, specifically That is, the ratio of the matching number of the rising and falling between user i and user j to the matching number of user i itself.

[0035] The clustering module is specifically used for:

[0036] Take each column of data in M as a sample to construct a degree matrix D:

[0037] Construct a Laplacian matrix L based on the degree matrix D: L = D - M;

[0038] Perform eigenvalue decomposition on the Laplacian matrix L, and take the eigenvectors corresponding to the first K smallest eigenvalues to form a matrix V, where K is the number of branches;

[0039] Perform K-means clustering on each row of the matrix V to divide the users into K clusters.

[0040] On the other hand, the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the low-voltage substation branch relationship verification method based on voltage record spectrum clustering between users as described in the first aspect.

[0041] On the other hand, the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the low-voltage substation branch relationship verification method based on voltage record spectrum clustering between users as described in the first aspect.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1) By means of the analysis method based on the recording of the user voltage change moment, the present invention can achieve high-precision identification without injecting additional power signals and only relying on the existing data. The algorithm design is efficient, the calculation period is short, and the branch relationship identification task can be quickly completed, meeting the real-time requirements of the dynamic changes of the low-voltage distribution network.

[0044] 2) Based on the analysis of the voltage data in the power consumption information acquisition system, this method does not require additional installation of hardware devices such as sensors or monitoring devices. Compared with the traditional method, the present invention significantly reduces the equipment installation and operation and maintenance costs required for the transparency of the distribution network, providing an economical and efficient solution for large-scale engineering applications.

[0045] 3) Through accurate branch relationship identification, the present invention optimizes the operation and management mode of the low-voltage distribution network, improves the level of digitalization and transparency of the distribution network. At the same time, it provides an important technical support for the intelligent operation and maintenance of the low-voltage distribution network, helping to improve the power supply reliability and management efficiency of the distribution network.

[0046] 4) The present invention can handle complex tree-shaped or ring network structures in the low-voltage distribution network and has strong robustness to data noise and dynamic changes of branch relationships. In practical applications, the identification effect is stable, the adaptability is strong, and it can meet the branch relationship identification requirements in different scenarios.

[0047] 5) The present invention shows good performance in practical applications and is suitable for large-scale promotion. Its core algorithm is simple and efficient, with both theoretical feasibility and engineering practicality, and does not require high equipment investment, having broad commercial application prospects and can significantly improve the intelligent level of the low-voltage distribution network. Detailed implementation manners

[0048] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] According to the network topology and circuit principle of the low-voltage distribution network, when a user uses electricity, its current increases and the effective value of its own voltage will decrease accordingly. At the same time, the large voltage drop generated by the electricity consumption in this phase will cause a small increase in the voltage of other phases, and this cross-phase transfer is more obvious in the same branch box. Therefore, the branch relationship verification of the low-voltage substation area can be realized based on this. The first aspect of the present invention provides a method for verifying the branch relationship of a low-voltage substation area based on the voltage record spectrum clustering among users, and the specific steps are as follows:

[0050] Step 1: Extract the moments when the effective values of the voltages of each user decrease and increase

[0051] Step 1.1: According to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation area, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value within 0.1 second is greater than or equal to 0.1 volt, and form a record of the voltage change moments of the users. The data format of the moment record is year-month-day-hour-minute-second, and the time is in 24-hour format.

[0052] Step 1.2: Screen all the records of the decrease of phase A and the increase of phase B among different users and the records of the decrease of phase A and the increase of phase B of the same user according to the recorded three-phase voltages of all users. Here, A, B, and C are the default phase sequences. The output records include: the time points that meet the conditions (in the format of year-month-day-hour-minute-second), and the corresponding user numbers.

[0053] Step 2: Record the matching numbers of the rise and fall between the effective values of the voltages across phases

[0054] Step 2.1: Select all the moment record data of each user to be verified within the same day. Record the rising records as Up.k and the falling records as Down.k, where k represents the moment record data of the kth user. The statistical matching number of the rise of phase B and the fall of phase A is completed based on the voltage change moment record generated in Step 1. By traversing the time point data of all users, cross-phase matching events are screened and counted. num(Up.i→Down.j) is the matching number of the rise of phase B and the fall of phase A between user i and user j. num(Up.i→Down.i) is the matching number of the rise of phase B and the fall of phase A of user i itself. The num(*) function is used to count the number of elements in the record set.

[0055] Step 3: Construct a voltage correlation matrix according to the normalized matching coefficient, and the matrix element values reflect the correlation strength of the voltage changes between users.

[0056] Step 3.1: Construct an n×n matrix M according to the number n of users whose branch relationship needs to be verified. All initial elements of M are 0.

[0057] Step 3.2: Assign different numbers to the users to be verified, numbered 1, 2, 3,..., n.

[0058] Step 3.3: Assign values to the elements in the voltage correlation matrix M according to the normalized matching coefficient.

[0059] Specifically That is, the ratio of the number of up and down matching numbers between user i and user j to the number of matching numbers of user i itself.

[0060] Step 4: Perform spectral clustering and output the clustering results

[0061] Step 4.1: Take each column of data in M as a sample and perform clustering using the spectral clustering method.

[0062] Step 4.2: Construct the graph Laplacian matrix:

[0063] Calculate the degree matrix D:

[0064] Construct the Laplacian matrix L: L = D - M

[0065] Step 4.3: Perform eigenvalue decomposition on the matrix L, take the eigenvectors corresponding to the first K smallest eigenvalues, and form the matrix V, where K is the number of branches.

[0066] Step 4.4: Perform K-means clustering on each row of V to divide the users into K clusters.

[0067] Step 5: Check the branch relationship

[0068] Step 5.1: Compare the clustering results with the system archives and screen out abnormal users.

[0069] Specific example: The electricity consumption data of 17 branches in 5 substations in a certain community in Nanjing were selected as experimental data to verify this method. The experimental data were the voltage change time records and 96-point voltage data within 15 consecutive days of a certain month. The acquisition method of the 96-point voltage data was: the initial sampling time of each user was the same and samples were taken every 15 minutes, and each user could obtain a total of 96 time-point voltage effective value data per day.

[0070] Test results of the branch relationship recognition algorithm

[0071]

[0072]

[0073] The branch relationships of 17 branches in a total of 5 substations included in the test set are respectively identified and compared with the prior topological structure. The correct rate of branch relationship identification is 98.39%. It can be seen from the above table that this method can accurately identify branch relationships and has good effects over multiple days with high stability.

[0074] An embodiment of the present invention also provides a low-voltage substation branch relationship verification device based on voltage record spectrum clustering among users, including:

[0075] A voltage change time extraction module, configured to extract the times when the effective values of the voltages of each user in the same substation decrease and increase;

[0076] A matching number statistics module, configured to count the number of up and down matches between the effective values of the voltages of each user to be verified across phases;

[0077] An association matrix construction module, configured to construct a voltage association matrix according to the normalized matching coefficient;

[0078] A clustering module, configured to perform spectral clustering by constructing user samples according to the voltage association matrix;

[0079] A verification module, configured to verify the branch relationship by comparing the clustering result with the system file.

[0080] The voltage change time extraction module is specifically configured to: according to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation, extract and record the corresponding times when the absolute value of the decrease and increase of its effective value is greater than or equal to 0.1 volt within 0.1 second, and form a record of the voltage change times of the users. The record includes: the time point that meets the condition and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format.

[0081] The matching number statistics module is specifically configured to: select all the records of the voltage change times of each user to be verified within the same day, record the record of the rise of phase B as Up.k, and record the record of the fall of phase A as Down.k, where k represents that the record data at this time is the kth user, and A, B, and C are the default phase sequences; num(Up.i→Down.j) is the number of matches between the rise of phase B and the fall of phase A of users i and j; num(Up.i→Down.i) is the number of matches between the rise of phase B and the fall of phase A of user i itself; the num(*) function is used to count the number of elements in the record set.

[0082] The association matrix construction module is specifically configured to:

[0083] Construct an n×n matrix M according to the number n of users for which branch relationship verification is required. All initial elements of M are 0;

[0084] Assign different numbers to the users to be verified, with the numbers being 1, 2, 3,..., n;

[0085] Assign values to the elements in the voltage incidence matrix M according to the normalized matching coefficient, specifically That is, the ratio of the number of up-down matching numbers between user i and user j to the number of matching numbers of user i itself.

[0086] The clustering module is specifically used for:

[0087] Take each column of data in M as a sample to construct the degree matrix D:

[0088] Construct the Laplacian matrix L based on the degree matrix D: L = D - M;

[0089] Perform eigenvalue decomposition on the Laplacian matrix L, and take the eigenvectors corresponding to the first K smallest eigenvalues to form the matrix V, where K is the number of branches;

[0090] Perform K-means clustering on each row of the matrix V to divide the users into K clusters.

[0091] On the other hand, the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the low-voltage substation branch relationship verification method based on user voltage recording spectrum clustering as described in the first aspect.

[0092] On the other hand, the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the low-voltage substation branch relationship verification method based on user voltage recording spectrum clustering as described in the first aspect.

[0093] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0094] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement without departing from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A method for checking the branch relationship of a low-voltage substation based on clustering of voltage recording spectra between users, characterized in that: Including: Extract the moments when the effective voltage values of users in the same substation area decrease and increase; Statistically calculate the up-down matching numbers between the effective voltage values of each user to be verified across phases according to the extracted moments; Construct a voltage correlation matrix based on the matching numbers according to the normalized matching coefficient; Construct user samples based on the voltage correlation matrix for spectral clustering; Compare the clustering results with the system archives to verify the branch relationship.

2. The method for checking the branch relationship of a low-voltage substation based on clustering of user-intermediate voltage recording spectra according to claim 1, characterized in that: Extract the moments when the effective voltage values of users in the same substation area decrease and increase. Specifically: According to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same substation area, extract and record the corresponding moments when the absolute value of the decrease and increase of its effective value within 0.1 second is greater than or equal to 0.1 volt, forming a record of the voltage change moments of users. The record includes: the time point that meets the conditions and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format.

3. The low-voltage substation branch relationship verification method based on clustering of user-intermediate voltage recording spectra according to claim 1, characterized in that: Statistically calculate the up-down matching numbers between the effective voltage values of each user to be verified across phases according to the extracted moments. Specifically: Select all the records of voltage change moments of each user to be verified within the same day. Record the record of the rise of phase B as Up.k, and record the record of the fall of phase A as Down.k, where k represents that the record data at this moment is the kth user's, and A, B, and C are the default phase sequences; num(Up.i→Down.j) is the up-down matching number of phase B rise of user i and phase A fall of user j; num(Up.i→Down.i) is the up-down matching number of phase B rise and phase A fall of user i itself; the num(*) function is used to statistically calculate the number of elements in the record set.

4. The low-voltage substation branch relationship verification method based on clustering of user-intermediate voltage recording spectra according to claim 1, characterized in that: Construct a voltage correlation matrix based on the matching numbers according to the normalized matching coefficient. Specifically: Construct an n×n matrix M according to the number n of users whose branch relationship needs to be verified. All initial elements of M are 0; Assign different numbers to the users to be verified, and the numbers are 1, 2, 3,..., n; Assign values to the elements in the voltage incidence matrix M according to the normalized matching coefficient, specifically That is, the ratio of the number of up-down matches between user i and user j to the number of matches of user i itself.

5. The method for checking the branch relationship of a low-voltage substation based on clustering of voltage recording spectra between users according to claim 1, characterized in that: Construct user samples based on the voltage correlation matrix for spectral clustering. Specifically: Take each column of data in M as a sample to construct the degree matrix D: Construct a Laplacian matrix L based on the degree matrix D: L = D - M; Perform eigenvalue decomposition on the Laplacian matrix L, and take the eigenvectors corresponding to the first K minimum eigenvalues to form a matrix V, where K is the number of branches; Perform K-means clustering on each row of the matrix V to divide the users into K clusters.

6. A low-voltage substation branch relationship verification device based on user-intermediate voltage recording spectrum clustering, characterized in that: Including: A voltage change moment extraction module for extracting the moments when the effective voltage values of users in the same substation area decrease and increase; A matching number statistics module for statistically calculating the up-down matching numbers between the effective voltage values of each user to be verified across phases according to the extracted moments; A correlation matrix construction module for constructing a voltage correlation matrix based on the matching numbers according to the normalized matching coefficient; A clustering module for constructing user samples based on the voltage correlation matrix for spectral clustering; A verification module for comparing the clustering results with the system archives to verify the branch relationship.

7. The low-voltage substation branch relationship verification device based on user-intermediate voltage recording spectrum clustering according to claim 6, characterized in that: The voltage change moment extraction module is specifically used for: according to the high-frequency recorded wave data of the three-phase actual voltage values of each user in the same transformer area, extracting and recording the corresponding moments when the absolute value of the decrease and increase of its effective value is greater than or equal to 0.1 volt within 0.1 second, forming the voltage change moment record of the user. The record includes: the time point that meets the conditions and the corresponding user number; the time point format is year-month-day-hour-minute-second, and the time is in 24-hour format.

8. The low-voltage substation branch relationship verification device based on user-intermediate voltage recording spectrum clustering according to claim 6, characterized in that: The matching number statistics module is specifically used for: selecting all the voltage change moment records of each user to be verified within the same day, recording the record of the B-phase rising as Up.k, and recording the record of the A-phase falling as Down.k, where k represents that the moment record data is the kth user's, and A, B, and C are the default phase orders; num(Up.i→Down.j) is the matching number of the B-phase rising of user i and the A-phase falling of user j; num(Up.i→Down.i) is the matching number of the B-phase rising and A-phase falling of user i itself; the num(*) function is used to count the number of elements in the record set.

9. The low-voltage substation branch relationship verification device based on user-intermediate voltage recording spectrum clustering according to claim 6, wherein: The association matrix construction module is specifically used for: According to the number n of users whose branch relationship needs to be verified, constructing an n×n matrix M, and all initial elements of M are 0; Assigning different numbers to the users to be verified, and the numbers are 1, 2, 3,..., n; Assign values to the elements in the voltage incidence matrix M according to the normalized matching coefficient, specifically That is, the ratio of the number of up-down matches between user i and user j to the number of matches of user i itself.

10. The low-voltage substation branch relationship verification device based on user-intermediate voltage recording spectrum clustering according to claim 6, characterized in that: The clustering module is specifically used for: Take each column of data in M as a sample and construct the degree matrix D: Constructing a Laplacian matrix L based on the degree matrix D: L = D - M; Performing eigenvalue decomposition on the Laplacian matrix L, taking the eigenvectors corresponding to the first K minimum eigenvalues to form a matrix V, where K is the number of branches; Performing K-means clustering on each row of the matrix V to divide the users into K clusters.

11. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the low-voltage substation branch relationship verification method based on voltage record spectrum clustering between users as described in any one of claims 1-5.

12. A non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the low-voltage substation branch relationship verification method based on voltage record spectrum clustering between users as described in any one of claims 1-5.

Images